Intelligent Recognition And Research On Geographic Elements In The Scanned Map And System Implement

The extraction and recognition of the geography information is the basis of geographic information system(GIS) and global position system(GPS). Automation vectorization has transcendent advantages and has become the main method to capture the digital information.The paper describes Two-Dimensional Wavelet transform and the principle of its application in multiscale edge detection. The paper also provides the method of obtaining and recognizing geographic information from chromatic electronic map. Simulation tests show that multiscale edge detection of images using wavelet transform is much better when compared with the conventional Canny operator and sobel operator.The idea of object-oriented is also introduced in this paper, the geographic elements in the map are classified into line, symbol and text, and line furthermore into bar, arc(circle) and curve. This method recognizes each class of graphic objects from their natural characteristics. Each graphic object is recognized directly in its entirety at the pixel level. To vectorize the bar and arc objects accurately, the character segment is introduced as the guidance of geometrical parameters. The raster image is progressively simplified by erasing recognized graphic objects to eliminate their interference with subsequent recognition. Moreover using operations of mathematic morphology, this paper revectorizes the recognized map, the system using the method shows the significant improvement in speed and recognition quality.The present vector research all is carried on in the CAD/CAM domain, this paper first time attempts to carry on the research in the ordinary color map, and the algorithm brought forward has feasibility after proof,, and realized an automated vectorization system—Auto Vector on the VC++ 6.0 environment, which has the certain theory significance and the application value in the geography information system domain.